Improving the Selectivity of PACE4 Inhibitors through Modifications of the P1 Residue

Article abstract of DOI:10.1021/acs.jmedchem.8b01381

Paired basic amino acid cleaving enzyme 4 (PACE4), a serine endoprotease of the proprotein convertases family, has been recognized as a promising target for prostate cancer. We previously reported a selective and potent peptide-based inhibitor for PACE4, named the multi-Leu peptide (Ac-LLLLRVKR-NH₂ sequence), which was then modified into a more potent and stable compound named C23 with the following structure: Ac-dLeu-LLLRVK-Amba (Amba: 4-amidinobenzylamide). Despite improvements in both in vitro and in vivo profiles of C23, its selectivity for PACE4 over furin was significantly reduced. We examined other Arg-mimetics instead of Amba to regain the lost selectivity. Our results indicated that the replacement of Amba with 5-(aminomethyl)picolinimidamide increased affinity for PACE4 and restored selectivity. Our results also provide a better insight on how structural differences between S1 pockets of PACE4 and furin could be employed in the rational design of selective inhibitors.

Full text of DOI:10.1021/acs.jmedchem.8b01381

Subscriber access provided by TULANE UNIVERSITY
Article
Improving the Selectivity of PACE4 Inhibitors
through Modifications of the P1 Residue
Vahid Dianati, Pauline Navals, Frédéric Couture, Roxane Desjardins,
Anthony Dame, Anna Kwiatkowska, Robert Day, and Yves L. Dory
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01381 • Publication Date (Web): 01 Dec 2018
Downloaded from http://pubs.acs.org on December 2, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Improving the Selectivity of PACE4
Inhibitors through Modifications of the P1
Residue
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†
†
‡
‡
Vahid Dianati, Pauline Navals, Frédéric Couture, Roxane Desjardins,
‡
‡
‡
†
Anthony Dame, Anna Kwiatkowska, Robert Day * and Yves L. Dory *
†
Institut de Pharmacologie de Sherbrooke, Département de Chimie, Faculté des
Sciences, Université de Sherbrooke, 3001, 12e Avenue Nord, Sherbrooke,
Québec, J1H 5N4 (Canada)
‡
Institut de Pharmacologie de Sherbrooke, Département de Chirurgie/Urologie,
Université de Sherbrooke, 3001, 12e Avenue Nord, Sherbrooke, Québec,
J1H 5N4 (Canada)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
PACE4, a serine endoprotease of the proprotein convertases family, has been
recognized as a promising target for prostate cancer. We previously reported a
selective and potent peptide-based inhibitor for PACE4, named the multi-Leu
peptide (Ac-LLLLRVKR-NH sequence), which was then modified into a more
2
potent and stable compound named C23 with the following structure; Ac-DLeu-
LLLRVK-Amba (Amba: 4-amidinobenzylamide). Despite improvements in both in
vitro and in vivo profiles of C23, its selectivity for PACE4 over furin was
significantly reduced. We examined other Arg-mimetics instead of Amba to regain
the lost selectivity. Our results indicated that the replacement of Amba with 5-
(
aminomethyl)picolinimidamide (Ampa) increased affinity for PACE4, and restored
selectivity. Our results also provide a better insight on how structural differences
between S1 pockets of PACE4 and furin could be employed in the rational design
of selective inhibitors.
ACS Paragon Plus Environment

Page 3 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
There is substantial evidence that proprotein convertases (PCs) are broadly
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
involved in the malignancy of tumors and angiogenesis. This family of serine
endo-proteases consists of seven members namely PC1/3, PC2, furin, PACE4,
PC4, PC5/6 and PC7, which process their substrates at the consensus motif of
2
R-X-R/K-R-↓-X, X being any amino acid residue. The role of PCs in malignancies
is through the activation proteolysis of oncogenic precursor proteins. Among these
substrates, growth factors and their receptors (e.g, transforming growth factor-β
and insulin-like growth factor-1 receptor family members) are crucial for cell
growth. Other PC substrates, such as proteases from ADAM (a Disintegrin and
metalloproteinase) and MMP (Matrix-metalloprotease) families, as well as adhesion
molecules (e.g. E-cadherin) are crucial for cell adhesion and metastasis.3
Additionally, the list of substrates includes other regulatory proteins, along with
bacterial and viral toxins.4
The cellular overexpression of PCs provides a clue for their role in tumorigenesis,
5
as observed in many malignant cell types. This is the case for PACE4 which is
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
overexpressed in prostate cancer (PCa) and other cancer cell lines.6-10 mRNA
silencing studies demonstrated that inhibition of PACE4 had effects on
tumorigenesis and neovascularization PCa cell line and animal models.11 More
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
recently, pro-growth differentiation factor-15 (pro-GDF-15) was identified as a
PACE4 specific substrate in PCa involved in the proliferative phenotype. A PACE4
isoform, known as PACE4-altCT is overexpressed in PCa cell lines has been
found to be responsible for sustained tumor progression.12 It is clear that PACE4
inhibition could open a new therapeutic strategy for PCa either as mono or co-
therapy, thus justifying our increased efforts to develop clinically relevant PACE4
inhibitors.
In a previous study, we showed that a lipophilic tail composed of four leucine
residues attached to the N-terminus of the RVKR tetrapeptide, was critical to
increase the selectivity of PACE4 inhibitors. Thus, our octapeptide, named multi-
Leu (compound 1), inhibits PACE4 and furin with K = 22 and 430 nM, respectively
i
(
Figure 1).13 The replacement of P1-Arg with 4-amidinobenzylamide (Amba) and
P8-Leu with its D isomer resulted in our current lead, named C23 (compound 2;
ACS Paragon Plus Environment

Page 5 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 1), with improved PACE4 affinity (K = 4.9 nM). Whereas the multi-Leu
i
peptide 1 was rapidly metabolized when tested in vivo, C23 was much more
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
stable and consequently displayed prominent pharmacological efficiency (IC50
=
2
5 and 45 µM for DU145 and LNCaP PCa cells, respectively) with rapid uptake
by xenografted tumors, and a human plasma half-life of 1.7 h.11, 14 However,
Amba develops stronger interactions with the furin S1 pocket than the C-terminal
Arg residue in compound 1, resulting in a significant reduction in selectivity (only
2
-fold) for PACE4 over furin. The furin preference for P1-Amba is even more
visible with the tetrapeptide Ac-RVK-Amba. This simple ligand is twice as selective
toward furin, emphasizing the undeniable role and necessity of the four leucine-
tail for PACE4 selectivity.15
Figure 1. Structure of control PACE4 inhibitors 1 and 2.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
In another study, we successfully improved the selectivity of the multi-Leu peptide
by introducing β-branched basic residues in the P3 position. This led to a 40-
fold selective inhibitor.16 However, further studies determined that this type of
compound, harboring four consecutive basic residues, was devoid of PCa cells
antiproliferative activity, suggesting a lack of cell penetration to reach the PACE4-
altCT intracellular target. Trials to improve the selectivity of C23 by manipulating
its P5-P8 portion met with limited success (3-fold selectivity in favor of PACE4).
All these data reveal that the Amba residue constitutes a barrier for achieving
more selective compounds.17 This work relates our efforts to find alternative
residues that could successfully replace Amba at the P1 position.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Homology models of PACE4 suggest that the S1 pocket, which accommodates
the P1 residue of the inhibitors, is constructed from two remote regions (primary
structure): The first one described as a -sheet then loop motif runs from S305
to G319 and the second one, a loop-helix-loop, encompasses residues S345 to
I364. These two parts of the cleft are clamped together by means of several
interactions including a Ca2+ cation involved in salt bridges with D310, D353,
ACS Paragon Plus Environment

Page 7 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
D358 and E381 (Figure 2).18 The guanidinium ion of the P1 residue (Ac-RVKR-
NH ) is strongly held inside the S1 pocket by means of ionic forces with
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
aspartates 310 and 358, as well as ion-dipole interactions with the carbonyl
groups of P308 and A344. These interactions are identical within all PCs.18
However, despite the high degree of homology in and around the S1 pocket,
differences exist, as observed in the matching sequences 300HDSCN and
352GDYCS of furin and PACE4, respectively (see Figure 2). One hypothesis is
that these disparities, that concern three residues only, alter the shape of the S1
pocket in furin and PACE4 and may be responsible for the observed selectivity
differences between compounds 1 and 2.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 2. Stereo representation of a PACE4 P1–P4 active site homology model
2
+
with docked Ac-RVKR-NH (orange) inhibitor. The Ca cation (green sphere)
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
located deep inside the S1 subsite is essential for its stability.
Generally, refining the selectivity is more laborious than enhancing potency,19 and
given the high degree of homology between furin and PACE4, discrimination
between these two enzymes is challenging. Although C23 is highly potent in
blocking tumor progression of xenograft PCa animal models, inhibition of the most
ubiquitous member of the PC family, furin, could potentially lead to unforeseen
side effects and drawbacks. In the present study, new residues are rationally
designed, then introduced in the P1 position, based on available structural data
and structure activity relationship (SAR) studies conducted on both PACE4 and
furin. The main goal of our investigations is to determine the structural factors
(
P1 position) that might discriminate between PACE4 and furin, in order to create
potent and selective PACE4 inhibitors.
RESULTS AND DISCUSSION
ACS Paragon Plus Environment

Page 9 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Design and binding affinities. Two groups of mimetics were designed (and
1
3, 20
biologically tested: PACE4 and Furin K, DU145 and LNCaP cell lines IC )
i
50
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
for the P1 residue of PACE4 inhibitors (Figure 3): a) an aliphatic series in which
inhibitors 3-6 possess a guanidine group like Arg itself; and b) an aromatic series
whose members 7-13 bear an amidine function like Amba. Compound 3 is an
epimer of 1 at position P8. Both compounds have similar K values, but 3 is
i
more selective (PACE4 K = 24 nM, 32 times selective for PACE4). The D residue
i
at position P8 imparts metabolic stability and this is likely the reason why 3 is
endowed with an improved ability to inhibit PCa cell proliferation. Accordingly, all
newly synthesized inhibitors discussed in this work have a DLeu residue at
position P8, like compounds 2 and 3.20 The first two derivatives 4 and 5 were
rigidified analogs of 3, from which the terminal amide had been removed
(agmatine). The alkene 4 proved to be much better than the alkyne 5.14 The
improved affinity of 4 (K = 13 nM), compared to compound 3, however was
i
associated with substantial reduction of selectivity (3-fold for PACE4). Conversion
of the C-terminal amide to an alcohol moiety in 6 resulted in a poor and non-
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
selective inhibitor of PACE4. These observations are, however, consistent with
the poor inhibition of peptide alcohols reported for other serine proteases.21-22
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Together with the Amba derivative 2, the compounds 7 and 8 were used to carry
out a preliminary SAR study of the aromatic amidine. Contrary to the Arg analogs
3-6, the side chain of Amba is more rigid and it is also bulkier.18, 23 It is
anticipated that the amidine group will be held in the S1 pocket by the same
enzyme residues that interact with the equivalent guanidine of Arg (Figures 2
and 5). As can be observed, P308 and A344 carbonyls, and D310 and D358
carboxylates bind tightly to the two external NH parts of the guanidinium ion.
2
Additionally, D310 can develop an extra interaction with the arginine delta-NH,
so any designed mimics ideally need an equivalent hydrogen bond donor.
First, the analogues 7 and 8 were designed to assess the width and length of
the S1 pocket. Further extension of Amba with one methylene unit, producing
inhibitor 7, not only did not offer any improvement in selectivity, but also reduced
the binding affinity (PACE4 K = 17 nM). The tetra-fluorinated Amba derivative 8
i
has very little affinity toward PACE4 and furin. There are two possible explanations
ACS Paragon Plus Environment

Page 11 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
for this negative result: Either the S1 pocket is too narrow to accommodate the
four fluorine atoms or the positive charge is lacking since the calculated pK of
a
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
the tetrafluoro-benzamidine is as low as 6.9.24
According to plans, O and NH substituents were introduced at the ortho position
of the benzamidine, inside a fused 5-membred ring as in 9 and 10 respectively,
then as free phenol 11 and aniline 12. Finally, the pyridine 13 was designed as
a more direct mimetic of 2, since it could fill up the same space in the S1
pocket. The inhibitor candidates 9–13 were synthesized then tested biologically.
The bicyclic systems 9 and 10 proved essentially inactive; whereas, the Ki values
of the corresponding open systems, 11 and 12, are in the same range as those
of multi-Leu 1 and its diastereomer 3, albeit with lower selectivities. Whereas,
compound 13 is indeed a strong and selective inhibitor (K = 2.6 nM, PACE4).
i
The P1 residue of compound 10 has been recently used in P1 of furin inhibitors
with no success.25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. Structure of P1 arginine mimetics used for PACE4 inhibitors with general
a
structure of Ac-DLeu-Leu-Leu-Leu-Arg-Val-Lys-NHR apart from 1 with Leu at
position P8 instead of DLeu.13 The inhibition of PACE4 and furin are represented
b
as K ± SD, and antiproliferative activity on PCa cell lines as IC ± SEM. Data
i
50
c
adapted from Ref. 14; Not calculable, indicates that the curve did not converged
d
to 50% with doses up to 150 µm; Not determined, due to solubility/precipitation
problems.
In order to provide a rationale for the various K values, that are linked to the
i
mode of binding of the P1 side chains, DFT calculations were run (Figure 4).26
These calculations were intended to disclose the minimum energy conformations
ACS Paragon Plus Environment

Page 13 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
of Amba derivatives and other relevant properties. The pKa figures of the
amidinium ions and ortho functional groups were also estimated.24 It was first
confirmed that the rigid Amba (side chain in Figure 4b) is indeed a good mimic
of Arg (Figure 4a). For these two P1 residues, the distances between the Cα
and the central cation C atoms are 6.24 Å and 5.78 Å for Arg and Amba
respectively (Figure 5). Amba, being slightly shorter, can fit in the S1 cavity
without much distortion. This does not hold true for the longer Amba analog 7,
as demonstrated by its lower affinity (17 nM). In terms of charges, both guanidine
and amidine are also similar since they exist as guanidinium and amidinium
cations at physiological pH. By incorporating these functional groups in rings
through addition of O and NH atoms, the resulting bicyclic systems (Figure 4c
and 4d) are fully aromatic. Consequently, their N atoms are no longer basic (pKa:
1.0-3.5) and are not positively charged at physiological pH. As for compound 8
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
mostly neutral at pH 7.4), which suffered the same drawback, the affinities of
analogs 9 and 10 are drastically diminished. Upon breaking the N-O and the N-N
bonds of their heterocycles, the amidine moiety becomes strongly basic (Figure
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4
e and 4f) and the corresponding peptidomimetics 11 and 12 recover some
inhibitory activity, the aniline 12 being much stronger than the phenol 11 (Figure
). Calculations show that at physiological pH (7.4), the phenolate anion is the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
predominant species. Since a negative charge in that region of the inhibitor may
lead to unfavorable interactions with the carboxylate of D310 (Figures 5), this
may easily account for the lower affinity of inhibitor 11 for both furin and PACE4.
In addition, the introduction of the aniline and phenol groups increases steric
hindrance in the deep S1 subsite and may explain why both 11 and 12 are less
potent than 2.
ACS Paragon Plus Environment

Page 15 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 4. Energy-minimized (DFT) side chain conformers of arginine (a) and
arginine mimetics (b, e-g) and estimated pK values of relevant functional groups.
a
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Torsion angles between amidine and aromatic planes are shown (θ).
Despite its additional NH group, the aniline (Figure 4f) is geometrically very
2
close to its parent amidine devoid of substituents (Figure 4b). For both cases,
the amidinium plane is rotated relative to the aromatic ring by ~ 35°. Surprisingly,
this torsion angle is smaller in the case of the aniline, because of its ortho
position that induces the formation of an intramolecular hydrogen bond N−H∙∙∙N.
However, for obvious steric reasons, the aniline-substituted amidine is prevented
from reaching a fully flat geometry, contrary to the simpler amidine that can, at
a cost of 2.8 kcal.mol−1 according to DFT calculations. The K values (PACE4)
i
for the corresponding inhibitors 2 and 12 are 4.9 nM and 14 nM, respectively
(Figure 3).
The best compound of the whole series is the pyridine 13, in terms of affinity
for PACE4 (2.6 nM) as well as selectivity (8 for compound 13 compared to only
2
for compound 2). This overall beneficial effect does not arise from a better
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
electronic complementarity with D310, because the pyridine nucleus is so electron-
deficient that it is not basic at all (pK : -1.6, Figure 4g) and remains neutral at
a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
all pHs. Nevertheless, the lone pair of the pyridine is ideally positioned for an
intramolecular hydrogen bond N∙∙∙H−N with the neighboring amidinium partner. As
a result, the terminal side chain of 13 is very flat indeed, its torsion angle  is
as small as 7°, a significant gain of 30° by comparison with isosteric peptide
mimic 2 (see Figure 4b and 4g). Induced-fit docking models (Figure 5) suggest
that the pyridine N atom makes no additional interaction with PACE4, in which
case its sole purpose is to freeze the amidinium side chain in its flat conformation.
From all these results, it can be inferred that a planar conformation of the P1
residue might fit better in PACE4 S1 pocket compared to furin.
ACS Paragon Plus Environment

Page 17 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 5. Superimposed induced fit docking pose of Ac-RVKX corresponding to
the P5-P1 region of compounds 1 (orange) and 13 (green) in the PACE4
homology model active site.18 Enzyme’s side chain C atoms colored the same
as corresponding ligand for clarity. H, N and O atoms are in white, blue and red
color, respectively. Hydrogen bonds are represented as yellow dashes.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Cell-based assays. In order to evaluate the cell antiproliferative activities, the
inhibitors were tested against PCa cell lines (DU145 and LNCaP) using an MTT
proliferation assay (Figure 3). In terms of cell-based efficacy, neither compounds
4
nor 5 showed significant advantage over compound 3 despite their differences
in PACE4 affinity and selectivity. However, the poor affinity of inhibitor 6 for
PACE4 entirely translated into antiproliferative activity, and no effect was observed
on both cell lines for this inhibitor in concentrations below 150 µM further depicting
the relationship between the anti-proliferative response and PACE4 inhibition.
Inhibitors 9 and 10, precipitated during the assay and, thus, their IC50 values
were not measurable.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. Cell permeability comparison of FITC-labeled analogues 14 and 15 of
compounds 2 and 13.
Analogues 7, 8, 11 and 12 which are less potent PACE4 inhibitors compared to
the parent compound 2, displayed inferior antiproliferative activity (Figure 3).
Whilst still active, inhibitor 13 (IC50 = 130 and 140 µM for DU145 and LNCaP
cells, respectively) displayed lower antiproliferative activity than control compound
2
(IC50 = 25 and 45 µM for DU145 and LNCaP cells, respectively). Knowing the
requirements of PACE4 inhibition to reach the intracellular PACE4 for efficient
antiproliferative activity, the cell permeability of 13 was tested using its N-terminally
FITC labelled version (compound 15).13,14,27 However, the results depicted no
ACS Paragon Plus Environment

Page 19 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
considerable difference in its permeability compared to the control FITC-labeled
analogue 14 corresponding to inhibitor 2 (Figure 6).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Overall, compound 13 with the best affinity for PACE4 and reasonable selectivity
among the P1 modified inhibitors provided reasonable antiproliferative activity in
PCa cell lines. The Ampa could be an appropriate P1 residue for further
development of more selective PACE4 inhibitors.
Chemistry. All the peptides 5 and 7−13 were synthesized using a combination
of solid and solution phase peptide synthesis as reported earlier for compound
2
.14 Briefly, the fully-protected P8-P2 peptide was synthesized with conventional
Fmoc-tBu strategy on 2-chlorotrytylchloride resin. After coupling of the last amino
acid and removal of the Fmoc protective group, the peptide was acetylated, then
cleaved from resin under mild acidic conditions. The P1 residue amines were
then coupled with the protected peptide in solution. The fragment coupling
procedure used to connect the heptapeptide to arginine mimetic synthons may
obviously lead to some degree of epimerization at the P2 position, the lysine
residue. However, the final products were proven to be pure by 1H NMR
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
spectroscopy after reverse phase preparative HPLC. Compound 6 was prepared
in a different manner using a literature method for the synthesis of C-terminal
alcohol peptides.28 In the final step, the side chains of all the peptides, were
deprotected using TFA cocktails. The FITC-labeling of compound 15 was carried
out as previously reported for compound 14.14
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The alkyne 18, P1 residue of compound 5, was obtained from two consecutive
Mitsunobu reactions on 2-butyne-1,4-diol 16 and a final phthalimide deprotection
(Scheme 1). For the P1 residues of 9 and 10, compound 19 was converted to
a dinitrile 20. The regioselective formation of 5-membered ring, was followed by
reduction of the remaining nitrile moiety to yield compounds 21 and 22. The
inhibitor 11 was then prepared by catalytic hydrogenation of inhibitor 9.
Scheme 1. Synthesis of P1 arginine mimetics 18, 21 and 22 for inhibitors 5,
and 10 a
9
ACS Paragon Plus Environment

Page 21 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
o
Reagents and conditions: (a) phthalimide, DIAD, PPh , THF, 0 C to rt, 16 h;
3
o
(b) N,N'-Di-Boc-guanidine, DIAD, PPh , THF, 0 C to rt, 16 h; (c) N H .H O,
3
2
4
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
o
CHCl /MeOH, 4 h; (d) Pd(PPh ) , Zn(CN) , DMF, 80 C, 6 h; (e) AcNHOH, K CO ,
3
3 4
2
2
3
o
DMF, 12 h; (f) BH -THF 1 M in THF, 0 C to rt, 6 h; (g) N H .H O, n-BuOH,
3
2
4
2
reflux, 16 h.
The synthetic pathways of four other P1 residues are outlined in Scheme 2.
Compound 25 was obtained by condensation of 4-formylbenzonitrile 23 with
nitromethane prior to consecutive reduction of the double bond and nitro group,
then Boc protection of the resulting amine giving compound 25. The nitrile group
of this compound was then converted to an amidine and the Boc protection was
removed to afford compound 26.29 Nitromethane was also employed in a S Ar2
N
reaction to prepare 28 from pentafluoro-benzonitrile 27. The nitro product 28 was
transformed to the Arg mimetic 29 by conversion of nitrile to amidine. The final
step of the conversion to 29 included a catalytic reduction which the nitromethyl
moiety was reduced to the aminomethyl as well. The benzylic alcohol 31 was
prepared by oxidation of the methyl substituent in the starting material 30. The
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
alcohol 31 was transformed into the corresponding tosylate, then azide, which
was finally converted to the Boc-protected amine 32 via a Staudinger reaction.
The nitrile of 32 was transformed into an amidine as before (25 to 26), during
which process the nitro group was also reduced to an aniline (step f in Scheme
2). Cleavage of the Boc group yielded the amine 33. The bromide in 34 was
replaced by a nitrile, then its ester was reduced to the benzylic alcohol 35.
Mitsunobu reaction on 35, using phthalimide as nucleophile, followed by hydrazine
opening of the resulting phthalimide led to the corresponding amine, that was
immediately protected as its t-butyl carbamate 36. From there, the same three-
step procedure used to prepare 26 from 25, and 29 from 28, was applied to 36
to obtain the amine 37.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 2. Synthesis of P1 arginine mimetics 26, 29, 33 and 37 for inhibitors
, 8, 12 and 13a
7
ACS Paragon Plus Environment

Page 23 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
o
Reagents and conditions: (a) MeNO , NaOH, MeOH/H O, <10-15 C, 15 min
2
2
o
then 5 M HCl; (b) Bu SnH, CH Cl , rt, 16 h; (c) Zn, HCl(aq), 65 C, 1h; (d)
3
2
2
o
(
Boc) O, K CO , THF/H O, 16 h; (e) NH OH.HCl, DIPEA, MeOH, 60 C, 16 h;
2
2
3
2
2
(f) Ac O, DIPEA, THF then 10% Pd/C, AcOH/MeOH, 35 psi H , 12 h; (g) Conc.
2
2
o
o
HCl(aq), MeOH, 0 C to rt, 1 h; (h) MeNO , TMG, -35 C, 5 min; (i) H IO , CrO ,
2
5
6
3
o
MeCN, 3 h; (j) i-BuOCOCl, NMM, THF, 0 C, 2 min then NaBH in MeOH, 30
4
min; (k) TsCl, Et N, DMAP, MeCN, 1 h; (l) NaN , NaI, DMF, 1 h; (m) PPh , H O,
3
3
3
2
o
THF, 16 h then K CO , (Boc) O, 16 h; (n) Pd(PPh ) , Zn(CN) , DMF, 100 C, 16
2
3
2
3 4
2
o
h; (o) NaBH , LiCl, MeOH, 2 h; (p) phthalimide, DIAD, PPh , THF, 0 C to rt,
4
3
1
6 h; (q) N H .H O, CHCl /MeOH, 3 h.
2
4
2
3
CONCLUSION
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
In conclusion, we developed new inhibitors for PACE4 through SAR studies of
the P1 residue. Our results emphasize structural differences in the S1 pocket
that can be used to discriminate between furin and PACE4. This could open a
novel avenue for achieving higher selectivity. Inhibitor 13 is the most potent and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
selective inhibitor in the series and displayed K = 2.6 nM which is 2-fold more
i
potent and 4-fold more selective than C23 (compound 2). Although the P1-Ampa
in 13 introduces the same stabilizing Van der Waals contacts as Amba in 2, the
former is held rigidly (intramolecular H bond) in a flat conformation, suggesting
that its gain in affinity for PACE4 might precisely results from this rigidifying
process. Despite similarities in structures and cell permeabilities, surprisingly,
compound 13 exhibited lower efficiency in PCa cell antiproliferative assay
compared to 2. Further studies may provide the knowledge to explain this
impairment, however, this significant improvement at the P1 position can now be
used in combination with other modifications at other positions.
EXPERIMENTAL
ACS Paragon Plus Environment

Page 25 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Chemistry. All chemical reagents and solvents were obtained from commercial
resources and used without further purification. Fmoc-protected amino acids and
coupling reagents, were purchased from ChemPep (Miami, FL, USA) or Chem-
Impex International (Wood Dale, IL, USA), and 2-chlorotrityl-chloride resin from
Rapp Polymer (Tübingen, Germany). Other reagents were purchased from Sigma
Aldrich (St. Louis, MO, USA). Manual Fmoc/tBu strategy was utilized for peptide
synthesis. Analytical reverse phase high-performance liquid chromatography (RP-
HPLC) was performed on an Agilent Technologies 1100 system equipped with a
diode array detector (λ = 210, 214, 230, and 254 nm). Preparative HPLC were
accomplished on a Waters preparative HPLC machine (Autosampler 2707,
Quaternary gradient module 2535, UV detector 2489 (λ = 214 and 230 nm),
fraction collector WFCIII) equipped with an ACE5 C18 column (250 × 21.2 mm,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
µm spherical particle size). Analytical HPLC was carried out using C18 columns,
either Agilent Eclipse XDB (5 mm, 4.6 µm, 250 mm) or a Phenomenex Jupiter
(5 mm, 4.6 µm, 250 mm). A gradient of H O/MeCN containing 0.1% TFA was
2
used as eluent for both analytical and preparative HPLC. The identity of the pure
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
products was confirmed using an ESI-HRMS system (TripleTOF 5600, ABSciex;
Foster City, CA, USA). For synthesis check a Water H Class Acquity UPLC
coupled with an SQ Detector 2 and a PDA eλ detector paired with an Acquity
UPLC CSH C18 column (1.7 µm × 2.1 mm × 50 mm) was used with a linear
gradient from 5 to 95% of MeCN containing 0.1% formic acid in 0.1% aqueous
formic acid was used for 1.3 min and a flow rate of 0.8 mL/min. NMR experiments
were carried out on either AV300 Bruker (300 MHz for H and 75 MHz for 13C)
or Avance III hd 400 Bruker (400 MHz for H, 377 MHz for 19F and 100 MHz
for 13C) or Agilent Varian (600 MHz for H) spectrometers. Complete decoupling
of protons was applied in 13C NMR experiments. The new compounds were
additionally characterized by IR spectroscopy (Alpha-Platinum ATR Bruker,
diamond crystal). The purity of biologically tested compounds 5−13 and 15 were
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
1
1
1
confirmed to be more than 95% using reversed phase analytical HPLC and H
NMR (13). The compounds were characterized by HRMS.
Peptide synthesis general procedures. Compounds 5 and 7−10, 12 and 13 were
prepared as follows; Fmoc-Lys(Boc)-OH (1.2 equiv) was attached to 2-chlorotrityl
ACS Paragon Plus Environment

Page 27 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
chloride resin in the presence of DIPEA (4 equiv) in DMF for 3h. The resin was
washed using a sequence of DCM (3 ×), a cycle of MeOH/DCM (3 ×) and DCM
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
5 ×). The peptide chain was grown by standard Fmoc-SPPS. The Fmoc group
was removed with 20% piperidine in DMF. The Fmoc protected amino acids were
coupled using 5 equiv of protected amino acids, 5 equiv of HATU, and 15 equiv
of DIPEA in DMF for 45 min. The last residue was acetylated by a solution of
Ac O/DIPEA/DCM (5:10:85) for 30 min. Each protected peptide was released from
2
the resin with a solution of 20% hexafluoro-2-propanol in DCM. The solvents
were evaporated in vacuo, and the residue was lyophilized in tert-BuOH/H O
2
(50:50). DIPEA (7.5 equiv) was added to an ice-cooled solution (0 °C) of protected
peptide, arginine mimetic (2.5 equiv), PyBOP (2.7 equiv) and 6-Cl-HOBt (7.5
equiv) in DMF and the reaction stirred overnight at room temperature. The solvent
was removed with an air stream to afford the crude protected peptide. The final
deprotection of side chains was achieved by a solution of H O/TIPS/TFA
2
(
2.5:2.5:95) for 2 h. Purification with reversed phase preparative HPLC (gradients
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
of H O/MeCN containing 0.1% TFA) resulted in pure peptides as their TFA salts
2
after lyophilization.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Compound 6. Fmoc-Arg(Pbf)-ol (0.63 g, 0.99 mmol)30 was added to a 2% solution
of DBU in DMF (7.0 mL) and agitated with 2-chlorotrytylchloride resin (0.3 g,
0.33 mmol) for 7 h. Fmoc-Lys(Boc)-OH (0.93 g, 1.98 mmol) was coupled to the
free alcohol using DIC (0.15 mL, 0.99 mmol) and DMAP (0.01 g, 0.11 mmol).
Then, the peptide was synthesized via deprotection/coupling cycles on resin as
previously mentioned in the general procedure for peptide synthesis. After final
acetylation, the protected peptide was obtained by treating the resin with 25%
HFIP in DCM (5.0 mL). Solvents were removed and the O-N acyl migration was
performed by stirring a solution of the resulting crude product in 20% piperidine
in DMF (4.0 mL) for 1 h. Global deprotection was carried out as mentioned in
the general procedure for peptide synthesis. The crude peptide was purified by
preparative HPLC as described in the general procedure.
Compound 11. The peptide inhibitor 9 (0.05 g, 0.05 mmol) was dissolved in
MeOH/H O (50:50, 4.0 mL). 10% Pd/C (50 mg) was added and the slurry was
2
ACS Paragon Plus Environment

Page 29 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
stirred under H (1 atm) for 16 h. After completion of the reaction (monitored by
2
UPLC-MS), the reaction mixture was filtered through a pad of diatomaceous earth.
The pad was washed with distilled water and the filtrate evaporated to dryness
with a stream of air. The residue was purified by reversed phase preparative
HPLC as described in the general procedure.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
Compound 13. H NMR (600 MHz, D O) δ (ppm): 8.64 (d, J = 1.8 Hz, 1H), 8.01
2
(
d, J = 7.9 Hz, 1H), 7.94 (dd, J = 7.9, 1.8 Hz, 1H), 4.55 (AB d, J = 15.6 Hz,
1
3
H), 4.49 (AB d , J = 15.6 Hz, 1H), 4.32-4.16 (m, 6H), 4.04 (d, J = 7.9 Hz, 1H),
.14 (m, 2H), 2.96 (quin, J = 6.1 Hz, 2H), 2.01 (m, 1H), 1.97 (s, 3H), 1.86-1.70
(
m, 4H), 1.70-1.49 (m, 16H), 1.45 (m, 1H), 1.37 (m, 1H), 0.92-0.80 (m, 30H).
N-(4-Phthalimido-but-2-ynyl)-N',N''-1,3-bis(tert-butyloxycarbonyl)guanidine
The first Mitsunobu intermediate product was obtained as a colorless solid in
2% yield using a reported procedure with all obtained spectra agreed with the
(17).
5
literature.31 1H NMR (400 MHz, CDCl ) δ (ppm): 7.88 (dd, J = 5.4, 3.0 Hz, 2H),
3
7
.75 (dd, J = 5.4, 3.0 Hz, 2H), 4.49 (s, 2H), 4.25 (br. s., 2H). 13C NMR (100
ACS Paragon Plus Environment